the ice age and mammal speciation in north america

Download The Ice Age and Mammal Speciation in North America

Post on 15-Jan-2017




0 download

Embed Size (px)


  • The Ice Age and Mammal Speciation in North AmericaAuthor(s): A. L. RandSource: Arctic, Vol. 7, No. 1 (Jun., 1954), pp. 31-35Published by: Arctic Institute of North AmericaStable URL: .Accessed: 16/06/2014 04:22

    Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .

    .JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact


    Arctic Institute of North America is collaborating with JSTOR to digitize, preserve and extend access to Arctic.

    This content downloaded from on Mon, 16 Jun 2014 04:22:42 AMAll use subject to JSTOR Terms and Conditions


    A. L. Rand*

    A number of effects of the Ice Age on life in northern North America have

    J^ been briefly reviewed by Flint (1952) in an earlier number of this journal. To these should be added another, the effect of glaciation on speciation of plants and animals. This aspect of the Ice Age in North America has received some attention in regard to plants and birds and here it is shown to have had an effect on some mammals.

    As I pointed out in discussing the effects of glaciation on birds (Rand, 1948) in continental areas, when a wide ranging species varies geographically and is divided into a number of subspecies, the usual pattern is for a gradual change from one subspecies to another, the width of the area of intergradation being wider or narrower depending on the suddenness of the change of ecological conditions. As long as these subspecies are in contact they inter- breed, there is gene flow between the populations, and they all remain one species. Geographical isolation by physical barriers seems necessary for a population to evolve further, to the species level. In this isolation the two parts of the population gradually change by random mutation toward a state in which interbreeding with other populations is impossible. Then, when the barriers disappear, or are surmounted, the two descendents of the one ancestral form meet. What happens now depends on how far the process of speciation has gone. If it has not gone to the point where biological discontinuity is complete, they meet each other as subspecies and interbreed. If biological discontinuity has been achieved and their ways of life have become different enough they invade each other's territory and live together in the same areas as sympatric species.

    In birds in northern North America there are a number of cases where forms meet that fit neither extreme. They have reached a point where they do not intergrade as ordinary subspecies, but their ways of life are still so similar that they are unable to invade each other's territory, and in some cases hybridization occurs, but only in a narrow zone. Examples are the gull complex of the Larus argentatus-thayeri-kumlieni-leucoptera complex; the large and small Canada geese of the Branta canadensis complex in the Arctic, and the spruce and Franklin's grouse (Canachites canadensis and C. franklini) that meet in the Rocky Mountains of British Columbia and Alberta. These phenomena correlate well with postulated effects of isolation by glaciers during the Ice Age: populations which had been isolated in refugia on the fringe of the ice met again with the retreat of the glacier, and the present zones of contact are therefore secondary. The refugia which seem to be

    Curator of Birds, Chicago Natural History Museum.


    This content downloaded from on Mon, 16 Jun 2014 04:22:42 AMAll use subject to JSTOR Terms and Conditions


    indicated are: one (or more) southeast of the ice; two (or more) southwest of the ice in the Rocky Mountain area; one (or more) in the Alaska-Bering Sea area; and one in the arctic islands (this need not have been permanent in location, but have been slowly moving and transient in any one locality).

    The correlation of bird data and postulated Ice Age effects makes it advisable to examine some mammal examples (basing classification on Ander- son, 1946),1 illustrating this idea:

    (a) A Bering Sea area and a southeast refugium The tundra shrew (Sorex tundrensis) of Alaska, northern Yukon, and

    northern Mackenzie is replaced to the south and east by the closely related saddle-backed shrew (S. arcticus), that ranges from northeastern British Columbia and central Mackenzie east and south to eastern Canada. This seems to indicate a Bering Sea refugium and one south and east of the ice. Later there was a subsequent spread, in which the southeastern member was the most active, and a secondary meeting in the same habitat. Though now without geographical barriers to separate them, the way of life of these two animals is apparently so similar that they have been unable to penetrate each other's habitat.

    ( b) A southeast and a southwest refugium The red squirrels provide an example that correlates with two refugia

    south of the ice. Tamiasciurus hudsonicus is widespread in the coniferous forests, from the Atlantic north to the limit of trees, and west to Alaska and western British Columbia including Vancouver Island. However, in south- west British Columbia along the coast the douglas squirrel (T. douglasii) takes its place, and when the two meet they do not intergrade. The ranges of two species of marten are somewhat similar: Martes americana is found in the coniferous forest from northeastern North America to Alaska and British Columbia but is replaced west of the coast mountains by Martes caurina. In these instances, too, the eastern form has been the more active in spreading over much the larger area.

    (c) A Bering Sea area and a southwest refugium The thinhorn sheep (Ovis dalli) in its various local colour phases ranges

    from Alaska and Yukon into extreme northern British Columbia, to be replaced to the south by the bighorn sheep (Ovis canadensis). The ranges nearly, if they do not quite, meet and there is no sign of intergradation between the two species.

    A similar pattern is evident in the ranges of the two closely related ground squirrels. The Columbian ground squirrel (Citellus columbianus), that ranges in the mountains of central British Columbia and central western Alberta southward, is replaced in northern British Columbia and northward by the closely allied arctic ground squirrel (Citellus parry ii including plesius as a

    XI follow this standard list of names though later taxonomic work, notably that of Rausch (1953), has suggested some changes for nomenclatural reasons or because of changes in species limits.

    This content downloaded from on Mon, 16 Jun 2014 04:22:42 AMAll use subject to JSTOR Terms and Conditions


    subspecies). The arctic ground squirrel, however, shows a change in distri- bution pattern that adumbrates the distribution of some of the following examples. Adapted to alpine and tundra conditions of the north it was not restricted to the mountains as was its southern relative, and spread across the tundra of the northern mainland to Hudson Bay.

    The relationships of the grizzly bears (genus Ursus) have to be worked out, along the lines indicated by Rausch (1953, pp. 96-7), but from my cursory observations there are three main types in the north: a small Barren-Ground type of interior Alaska, Yukon, and the mainland Barrens of the Northwest Territories; a huge coastal Alaskan type (including the big brown bears); and the large bears of the Rocky Mountains and plains. If this be the case, it would fit the picture of two refugia in the Alaska area; and one refugium (or more) southwest of the ice. The picture farther south may be more complex. Apparently there was a large grizzly on the plains of Canada at least, but it was the small Yukon type, adapted to northern conditions, that spread east onto the Barrens.

    (d) A Bering Sea area and a southwest and southeast refugium A similar picture, but involving three refugia, is shown by the genus

    Microtus. The three species: meadow mouse (M. pennsylvanicus), tundra mouse (M. operarius), and Townsend mouse (M. toivnsendii) are largely representative species that do not intergrade where they meet or overlap. The meadow mouse ranges from the northeastern United States and eastern Canada to the Cascades of British Columbia, to be replaced in the extreme west by the Townsend mouse. To the north the meadow mouse meets the tundra mouse in central Yukon, where the latter replaces it at higher altitudes. In the Arctic farther east, the meadow mouse extends to the wooded mouth of the Mac- kenzie, and in Keewatin extends onto the southern Barrens, but on the Barrens fringing the Arctic Ocean at Coronation Gulf it is the tundra mouse that has won the race to occupy this area, while neither as yet seems to have occupied the Barrens of northern Keewatin.

    The red-backed mice of the genus Clethrionomy s also have three repre- sentative species that nearly if not quite meet, but without intergradin